, 2000; see also Voss et al., 2006). Anatomically, prediction of recovery from coma relies on the comprehensive assessment of all structures involved in arousal and awareness functions, namely, the ascending reticular activating system located in the postero-superior part of the brainstem and structures encompassing thalamus, basal forebrain, and fronto-parietal association cortices (Tshibanda et al., 2009). Lesion or inhibition of part of this system suffices to cause immediate coma (e.g., Parvizi and Damasio, 2003). Studies on traumatic coma patients

with conventional MRI showed that lesions of the pons, midbrain, and basal ganglia were predictive of poor outcome especially when they were bilateral (Tshibanda et al., 2009). In relation with the GNW model, it is noteworthy that prediction of nonrecovery after 1 year could be calculated with up to 86% sensitivity and 97% specificity when ZD1839 cost taking into account both diffusion tensor and spectroscopic measures of Screening Library screening cortical white matter integrity (Tshibanda et al., 2009). The objective neural measures of conscious processing demonstrated earlier in this review should be applicable to the difficult clinical problem of detecting consciousness in noncommunicating patients. Using fMRI, a few patients initially classified as vegetative by clinical signs showed essentially normal

activations of distributed long-distance cortical networks during speech processing and mental imagery tasks (Owen et al.,

2006 and Monti et al., 2010), and one patient proved able to voluntarily control them to provide yes/no answers to simple personal questions, clearly indicating some degree of preserved conscious processing (Monti et al., 2010). In an effort to isolate a more theoretically validated scalp signature of conscious sensory processing, Bekinschtein et al. (2009a) recorded ERPs to local versus global violations of an auditory regularity. When hearing a deviant tone after a sequence of repeated standard tones (sequence XXXXY), a local mismatch response was elicited nonconsciously even in coma and vegetative-state patients, as previously demonstrated (e.g., Fischer et al., 2004). However, when of this sequence XXXXY was repeatedly presented, such that the final tone change could be expected, the presentation of a deviant monotonic sequence (XXXXX) engendered a P3b wave in normal subjects that was absent in coma patients and in most vegetative-state patients but could still be observed in minimally conscious and locked-in patients. This paradigm, founded upon previous identification of the P3b component as a signature of conscious processing, is now undergoing validation as a means of identifying residual conscious processing in patients (Faugeras et al., 2011). The present review was deliberately limited to conscious access. Several authors argue, however, for additional, higher-order concepts of consciousness.